The proposed research will examine a web-based personalized feedback intervention to reduce alcohol-related risky sexual behavior among underage young adult drinkers aged 18-20, which is theoretically informed by the Prototype Willingness Model. Based on literature focusing on developmentally appropriate health models for adolescents and young adults, the Prototype Willingness Model assumes two pathways to risk behavior: one pathway that is reasoned and one pathway that is socially based. Prior research has shown that interventions consistent with the Prototype Willingness Model, or interventions based on components of the Prototype Willingness Model, are efficacious at reducing alcohol use and risky sexual behavior. Despite the utility of this model, it has yet to be applied to examining whether interventions comprised of components from both pathways of the PW Model are efficacious at reducing alcohol-related risky sexual behavior in underage young adult drinkers. Moreover, no research has examined the active components (reasoned pathway vs. social pathway) of this model. The research proposed in this application is designed to evaluate a personalized feedback intervention, based on the Prototype Willingness Model, focused on alcohol-related risky sexual behavior through a longitudinal study in a national sample of young adults (college and non-college) recruited via social networking sites. To accomplish this objective the study will enroll a national sample of 1,200 young adults aged 18-20 and assess them at 3-, 6, 9-, and 12-months. We will evaluate the overall efficacy of the interventions based on the Prototype Willingness Model by comparing underage young adult drinkers randomly assigned to receive the reasoned-based pathway intervention (n=300), the social-based pathway intervention (n=300), or the integrated intervention based on the full PW Model (both pathways; n=300) to an attention control group (n=300). We will examine whether changes in components of both the reasoned and social pathways and drinking mediate intervention efficacy on reducing alcohol-related risky sexual behavior. Past behavior and college student status will be evaluated as moderators of intervention efficacy. The proposed study is both significant and innovative in that it will evaluate brief interventions among a national sample of young adults attending and not attending college, will utilize social networking sites for participant recruitment, and will test the efficacy of interventions based on individual and integrated pathways of the Prototype Willingness Model.